Dynamic change-point detection using similarity networks

TL;DR

This paper introduces a method for detecting change-points in sequences of similarity networks by analyzing node-wise similarity measures, with applications in anomaly detection and faulty sensor isolation.

Contribution

It proposes a simple sequential change detection procedure based on node similarity, with theoretical analysis and practical validation on real and simulated data.

Findings

The method effectively detects change-points in network sequences.

Simulation and real-data examples show good detection performance.

Community detection helps isolate anomalous nodes.

Abstract

From a sequence of similarity networks, with edges representing certain similarity measures between nodes, we are interested in detecting a change-point which changes the statistical property of the networks. After the change, a subset of anomalous nodes which compares dissimilarly with the normal nodes. We study a simple sequential change detection procedure based on node-wise average similarity measures, and study its theoretical property. Simulation and real-data examples demonstrate such a simply stopping procedure has reasonably good performance. We further discuss the faulty sensor isolation (estimating anomalous nodes) using community detection.